Repository navigation
AI Visibility
Site → Visibility tab. Does your site get cited when real people ask AI assistants questions? This is "Answer Engine Optimization".
The tab has these sub-tabs:
| Sub-tab | Question it answers | Cost |
|---|---|---|
| AI answers | For my questions, asked live today: am I cited, mentioned or absent, per engine? | Paid (your AI keys) |
| LLM Mentions | What have models actually been answering about my brand, including questions I never tracked? | Paid (DataForSEO) |
| Share of voice | How often am I named vs. my competitors, and which domains do AIs cite instead of me? | Free (replays stored answers) |
| Mentions | Was my brand mentioned in the news, on Wikipedia or Wikidata, with or without a link? | Free |
| Yandex AI | Which queries' Yandex AI answers already cite my site? | Keys.so credits |
You add questions. OpenGSC asks each engine with live web search on, because an answer from a model's weights says nothing about search visibility.
| Engine | How it's asked | Key |
|---|---|---|
| ChatGPT | OpenAI Responses API + web_search
|
OpenAI key (Settings → AI models) |
| Claude | Anthropic Messages API + server-side web search | Anthropic key (Settings → AI models) |
| Gemini |
generateContent + Google Search grounding |
Gemini key (Settings → AI models) |
| Perplexity | Sonar (search built in) | Perplexity key (Settings → AEO & site health) |
| Grok | xAI Live Search | xAI key (Settings → AEO & site health) |
| Google AI Overviews | Your question goes through a real Google SERP; the overview's sources become citations | DataForSEO (SERP Advanced) or A-Parser |
You only need keys for the engines you want. The others are shown with a hint about what's missing.
Verdicts are three-state: cited (linked), mentioned (named, no link) and absent. AI Overviews has a fourth, no_overview: Google showed no overview for that question, which isn't the same as "not cited" and doesn't count in share-of-voice denominators.
The check shows its work. Expand a question to see the full answer, every cited domain (yours highlighted), your rank among them, the model that answered and whether a live search actually happened. If ChatGPT in another tab tells you something different, you can compare.
"Cited instead of you" lists the domains that keep appearing across your questions: the pages your answer has to displace. Each is classified into 12 categories (forums, social, video, reviews, reference, e-commerce, …) and page types (listicle, how-to, comparison, review…), with a dedicated Reddit & communities cut. The classifier is deterministic and free. An "other: NN%" chip warns when many citations go unclassified; extend the Domain lists (in the AI answers settings) with domains for your market. A Greek starter set ships built in.
Per-site settings: model, country, city and answer language. Location matters because local answers are geolocated.
Sentiment: for answers where you're cited or mentioned, one cheap model call rates the tone (positive / neutral / negative / mixed). It's off by default; run it with a button (price shown first) or turn on auto-assessment.
Question ideas: question-shaped queries from your own Search Console data, minus what you already track. Free.
Each question costs several billed calls on your keys per check (about four with all engines on). Automatic daily checking is off by default, so a big portfolio never enrols itself. Check all asks for confirmation.
A second source with a different job. The live tracker asks your questions on your keys today. LLM Mentions reads DataForSEO's index of what ChatGPT and Google AI Overview have actually been answering:
- how often the brand comes up and in which questions;
- which of your pages get cited;
- share of voice against up to 9 competitors.
Match by domain (answers that linked to you) or by brand name (answers that named you without a link). The index refreshes roughly monthly and covers ChatGPT and AI Overview only. A zero here isn't evidence of invisibility in Claude or Perplexity. The two panels stay side by side and labelled rather than merged.
Built from answers you already paid for, so it's free:
- how often each engine names you, your citation share, per-engine numbers and a weekly trend;
- up to 10 competitors. Adding a competitor recomputes the whole history at no cost;
- the domains AI actually cites for your questions, ranked. Each one is one click from your Outreach Workspace;
- question ideas mined from GSC.
Checks Google News, Wikipedia and Wikidata daily for your brand terms. These are free public sources. Add context words for ambiguous brand names (mustInclude). A mention without a link is the cheapest link-building lead there is: one click expands the news redirect, checks for a link and files the row into the Outreach Workspace. A daily summary can go to your notification channels (event mention).
SEO Tools → GEO Audit is a one-off deep dive into one query: which brands and source types an AI's search trace surfaces and why. The Visibility tab is ongoing tracking of many questions.
get_aeo_visibility, get_ai_share_of_voice, get_brand_mentions, get_yandex_ai_answers. All are free reads of stored data.
OpenGSC · opengsc.org · Repository · Issues · Changelog · MIT licence
Get started
Install and run
Analytics
- Dashboard
- Site Page
- Portfolio Analytics
- Rank Tracker
- AI Visibility (AEO)
- Indexing
- Audit, Health, Uptime
- Backlinks
Providers
Tools
Delivery
Help